fix(ported_static): bump gas budgets for EIP-8037 state-gas headroom on Amsterdam

[leolara]

🗒️ Description

Make tests/ported_static/ pass on Amsterdam under EIP-8037's two-dimensional gas model. Fix categories:

1. Gas-bump tests (OoG headroom): hardcoded tx_gas / inner-CALL gas in tests authored against pre-EIP-8037 budgets is too tight on Amsterdam (EIP-8037 splits SSTORE-set / CALL-with-value-to-new-account / SELFDESTRUCT-to-new-account into a smaller regular portion plus state-gas; when the reservoir is empty the state-gas spills back into regular gas). Each affected literal is replaced with a fork-conditional variable that keeps the original value on pre-EIP-8037 forks and uses a higher bump on Amsterdam.

2. Gas-measurement tests (Op.GAS before/after a state-changing op, value stored as post-state): the observed regular-gas consumption changes by deterministic deltas under EIP-8037:

   • fork.sstore_state_gas() - 17 100 per fresh SSTORE-set spill (17 100 = the SSTORE-set base regular drop)
   • fork.create_state_gas() - 25 000 per NEW_ACCOUNT spill (25 000 = old CALL_NEW_ACCOUNT / SELFDESTRUCT base)
   • sum of both for SELFDESTRUCT-with-write paths

   Expected storage values become original_value + (delta if fork.is_eip_enabled(8037) else 0). The fork-helper expressions auto-adapt to any future COST_PER_STATE_BYTE change (this branch already saw 1 174 → 1 530 mid-flight without further hand-tuning).

3. stCallCodes / stCallDelegateCodes* family (35 files): all share the same nested CALL/CALLCODE/DELEGATECALL pattern with three hardcoded inner gas values (0x3D090 / 0x493E0 / 0x55730). Replaced with fork-conditional inner_call_gas / middle_call_gas / outer_call_gas variables across the family.

Per review feedback (#2796 by @marioevz), every gas change preserves the original value as the else branch of if fork.is_eip_enabled(8037), so behavior on Cancun / Prague / Osaka is unchanged.

Stats

	Count
Test files modified	52
Skip-list entries removed	165 (× 3 fixture variants = 495 test runs unskipped)
Skip list before (bal/7)	1 062
Skip list after	897

Verification

uv run fill -m "not slow" --fork Amsterdam tests/ported_static/  →  16 482 passed, 0 failed, 2 448 skipped
uv run fill -m "not slow"                 tests/ported_static/  →  60 476 passed, 0 failed (Cancun/Prague/Osaka/Amsterdam)

Files by category

Gas-bump (fork-conditional tx_gas / inner CALL gas):

File	Pre-8037 → Amsterdam
stCreate2/test_create2collision_selfdestructed.py	tx 400 000 → 1 000 000, inner CALL 0xC350 → 0x40000
stCreate2/test_create2collision_selfdestructed2.py	tx 400 000 → 1 000 000, inner CALL 0xC350 → 0x40000
stCreate2/test_create2_smart_init_code.py	tx 400 000 → 1 000 000
stCreate2/test_create2_contract_suicide_during_init_then_store_then_return.py	tx 600 000 → 5 000 000, inner CALL 0x249F0 → 0x100000
stCreateTest/test_create_transaction_call_data.py	tx 100 000 → 500 000
stCreateTest/test_create_transaction_high_nonce.py	tx 90 000 → 500 000
stCodeSizeLimit/test_codesize_valid.py	tx 15M → 40M
stRevertTest/test_revert_opcode_in_init.py	tx 160 000 → 800 000
stSStoreTest/test_sstore_change_from_external_call_in_init_code.py	(bal/7's intrinsic-derived fix; this branch defers)
stInitCodeTest/test_call_the_contract_to_create_empty_contract.py	tx 100 000 → 500 000
stWalletTest/test_day_limit_construction.py	tx [817 083, 1 217 083] → [5 000 000, 7 000 000]
stWalletTest/test_wallet_construction.py	tx [1 225 023, 1 825 023] → [8 000 000, 10 000 000]
stHomesteadSpecific/test_create_contract_via_transaction_cost53000.py	tx 100 000 → 500 000, sender 0xF4240 → 0x4C4B40
stPreCompiledContracts2/test_call_ecrecover_overflow.py	tx 100 000 → 500 000 (inner ecrecover CALL 0xBB8 unchanged — that's the test premise)

Gas-measurement (fork-conditional expected-value deltas via fork helpers):

File	Delta added on Amsterdam
stEIP2930/test_storage_costs.py	+ fork.sstore_state_gas() - 17 100 to 10 SSTORE-set measurements
stEIP2930/test_varied_context.py	SSTORE-set, NEW_ACCOUNT, and SELFDESTRUCT-with-write deltas to 28 measurements
stEIP2930/test_manual_create.py	SSTORE-set delta to 2 measurements
stPreCompiledContracts/test_precomps_eip2929_cancun.py	NEW_ACCOUNT delta to 2 expect-entries (87 parametrizations)

stCallCodes / stCallDelegateCodes* (35 files): uniform inner_call_gas / middle_call_gas / outer_call_gas fork-conditional refactor.

✅ Checklist

• All: Ran fast static checks to avoid unnecessary CI fails, see also Code Standards and Enabling Pre-commit Checks:

  just static

• All: PR title adheres to the repo standard - it will be used as the squash commit message and should start type(scope):.
• All: Considered updating the online docs in the ./docs/ directory.
• All: Set appropriate labels for the changes (only maintainers can apply labels).
• Tests: Ran mkdocs serve locally and verified the auto-generated docs for new tests in the Test Case Reference are correctly formatted.
• Tests: For PRs implementing a missed test case, update the post-mortem document to add an entry the list.
• Ported Tests: All converted JSON/YML tests from ethereum/tests or tests/static have been assigned @ported_from marker.

Cute Animal Picture

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (devnets/bal/7@bcc2a87). Learn more about missing BASE report.

Additional details and impacted files

@@               Coverage Diff                @@
##             devnets/bal/7    #2796   +/-   ##
================================================
  Coverage                 ?   85.59%           
================================================
  Files                    ?      630           
  Lines                    ?    39606           
  Branches                 ?     3937           
================================================
  Hits                     ?    33899           
  Misses                   ?     5083           
  Partials                 ?      624           

Flag	Coverage Δ
unittests	85.59% <ø> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[marioevz]

Can we create a call_gas variable that has a different value depending on whether EIP-8037 is enabled in the fork where the test is being executed?

call_gas = 0x249F0
if fork.is_eip_enabled(8037):
  call_gas = 0x100000
# { (CALL 150000 0xc94f5374fce5edbc8e2a8697c15331677e6ebf0b 1 0 0 0 32) (SSTORE 1 (MLOAD 0)) }  # noqa: E501
contract_0 = pre.deploy_contract(  # noqa: F841
code=Op.POP(
Op.CALL(
                gas=call_gas,

This will allow us to preserve the original gas value in case we need to do further modifications, and it also makes it more clear why this gas bump was done in the first place.

[marioevz]

Same comment for the rest of the tests.

[spencer-tb]

Feel free to rebase on the latest devnets/bal/7! :)

[leolara]

@marioevz @spencer-tb rebased, added more tests fixes and did as Mario said with the conditional gas